But the file came wrong, all the subtitles were late! The actor said "You can't revenge him!", and only after some seconds the subtitle appeared.
I thought: "I could search for another subtitle file on the web, but why if I can just solve it with a small program?"
So there was my challenge: Make the subtitles fit, because I really wanted to watch that episode.
As the blog's spirit is to dive ourselves in programming adventures, I decided to use a language that I'm not that familiar with: Python.
The first thing I did was understand my problem. I needed to process a .srt file, and then modify it to fit my needs (in other words, make the subtitles a little earlier, or, more easily, N seconds earlier).
The "anatomy" of a .srt file is:
id
HH:MM:ss,mmmm --> HH:MM:ss,mmmm
Subtitle goes here. In our case, an exemple could be:
Você nunca se vingará!
where id = Integer value to identify the subtitle; HH = Hours; MM = Minutes; ss = Seconds and mmmm = Milliseconds.
Basically, it shows, for each subtitle, its identification, starting time, ending time and content.
What I needed to do? Create an algorithm that catches all the time lines and modify them to be N seconds earlier.
There's two things to discover: How to make I/O operations using Python and how to handle datetime objects. That wasn't so difficult, I found all the answers at Python's doc page: http://docs.python.org/tutorial/inputoutput.html.
The pseudo-code:
void Shift_Subtitles (file, seconds):
timedelta shift <-- timedelta(seconds)
content <-- file.read()
Foreach line in content:
If line is timeline:
get start_time and end_time
transform to datetime object
new_start <-- start_time - shit
new_end <-- end_time - shift
replace line with new_start and new_end
file.write(content)
And that's it! A solution for the subtitle problem using a simple algorithm. If you want to see the Python implementation, here it is:
import sys
import datetime
from datetime import timedelta
def shiftSubtitles(str_file, sec):
file = open(str_file, 'r')
lines = file.readlines()
file.close()
file = open(str_file, 'r')
content = file.read()
file.close()
shift = timedelta(seconds = int(sec))
for line in lines:
if '-->' in line:
# Here we get HH:MM:SS:ss
_start = (line.split('-->')[0].rstrip().lstrip()).replace(',', ':').split(':')
_end = (line.split('-->')[1].rstrip().lstrip()).replace(',', ':').split(':')
start = datetime(1999, 01, 01, int(_start[0]), int(_start[1]), int(_start[2]), 0)
end = datetime(1999, 01, 01, int(_end[0]), int(_end[1]), int(_end[2]), 0)
new_start = (start - shift).isoformat(' ').split(' ')[1].replace('.', ',')
new_end = (end - shift).isoformat(' ').split(' ')[1].replace('.', ',')
new_line = new_start + ',000 --> ' + new_end + ',000\n'
content = content.replace(line, new_line)
file = open(str_file, 'w')
file.write(content)
file.close()
def main(args):
shiftSubtitles(args[0], args[1])
if __name__ == '__main__':
main(sys.argv[1:])
The way to use it is very simple too:
$ > python shift_sub.py [filename].srt [seconds]
Well, I'm not a Python programmer, so I don't know if this is the best way to do it. If you know how to make this code better, please comment for us!
